Myths and Reality of the Internet of Things
The Internet of Things is an idea that has long been mythicized and has sometimes stoked irrational fears. It refers to the wide-ranging concept of interconnected objects for general as well as professional use. Basically, these objects can be any type of device equipped with sensors that collect usage data and transmit them over the Internet. Some of these objects are capable of reacting to their environment, based on the data collected by their internal programme or received over the network (remotely controlled). Finally, they can be connected to one another through the Internet or a local, often wireless, network, depending on the distance between them.
Nowadays, the spectre of the Internet-enabled domestic toaster remains a bit of a joke; that said, the Internet of Things has indeed become a reality and practically no applications, especially industrial applications, have escaped it. In the very near future, tens of billions of objects will be connected to the network, generating a mass of data that defies the imagination. It is estimated that in South Korea, the most connected country in the world, there are over 38 connected objects per 100 people, while in Canada, the number sits at about 12 (2015 figures of the OECD).
In future, many mundane day-to-day objects will generate a constant flow of data that will be monitored and analyzed by computers. This data will feed the development of new concepts such as smart cities, enable ubiquitous computing and power ambient intelligence, a watershed in technological evolution.
1. Communicating Objects Will Blow up the Internet!
The risk of overwhelming Internet infrastructure is a legitimate concern, but the weakest links have been identified and mitigation measures have been put in place. The main risk is due to Internet addressing, since each one of the connected devices needs a unique Internet address. It has been estimated that by 2020, 25 to 30 billion connected objects will require a unique address; however, since the classic Internet environment (IPv4) can only accommodate a maximum of 4.2 billion IP addresses, we are running straight towards a crisis of epic proportions. However, this gridlock is close to being resolved thanks to the rollout of the new IPv6 Internet protocol, with its virtually unlimited number of potential IP addresses (2128, i.e. 340 undecillion and then some).
In any case, not all connected objects will be directly linked to the Net; some of them will have their own infrastructure and local network, whether on a domestic scale (home automation, or smart houses) or on a municipal scale (smart cities), which will ease some of the pressure on collective infrastructure. Some local networks will be quite restricted, such as wireless personal networks (WPAN, Wireless Personal Area Network), used to connect an Apple Watch to an iPhone for example.
2. My Toaster Will Be Hacked!
Maybe, but not likely. The scary thing about the Internet of Things is that attacks that once upon a time only had virtual consequences can now have tangible, immediate effects, such as carbonized toast. Even worse, there was the recent hacking of Jeep Cherokees, where hapless drivers found that their air conditioning, radio, windshield wipers, acceleration and braking were taken over and remotely controlled. In July, following this demonstration, Chrysler had to recall over 1.4 million vehicles.
All technical innovations have chinks in their armour, and this type of application is still new. The important thing to remember is that security is critical to any connected object project, regardless of its scope, and security should never be taken for granted. Of course, potential security concerns should not cause a project to grind to a halt: if we had let ourselves be ruled by security fears in the past, we probably wouldn’t have trains, planes, computers or the Internet today. It’s a question of balance: as a rule, IT projects should be developed prudently, professionally, and with an eye to security.
3. My Home Appliances Will Spy on Me!
The idea that the proliferation of sensors, whether at work, at home or in public places, will allow your privacy to be invaded is not far-fetched. This concern is directly related to that of security, but is much vaster in scope, since your data can attract attention from more than « just » hackers: businesses, organizations, and the government, for instance. Indeed, personal data produced by connected objects can be irresistible to insurance companies, for example, who could offer rebates in exchange for installing a tracking device on your car that will provide them with data on your driving style. In this case, you would be consenting to giving up some of your privacy in exchange for a discount — which is not a problem, as long as you are aware that you are being watched and that you freely consent to this.
That said, these threats to privacy are not new and in fact predate the advent of connected objects in countless fields. The point is to not fall in the twin traps of paranoia and complacency. After all, the whole idea of connected objects is to enhance the quality of our lives, our environment, and our health, among other things. On the other hand, no technology is risk-free. Ask yourself this: would you actually swear off the Internet because your privacy might be at risk? Of course not! Just as Internet and Web open standards have created untold sociocultural and economic opportunities, so will connected objects. And the Internet of Things hasn’t actually created new privacy protection issues; it has merely magnified a pre-existing problem with increased masses of data.
4. Objects that Talk to Each Other Are Kind of Scary.
No, your connected fridge is not going to go berserk and instruct a nuclear reactor to melt down. Communication between connected objects is exceptional and usually limited to devices of a similar nature or use. Such objects usually have in common their communication protocol (TCP/IP) and exchanged data formatting (XML), and not much else. In fact, they don’t usually speak the same language since their functions are too different. In fact, the very expression The Internet of Things is deceptive, since it’s not a network as such but rather a collection of highly specialized devices with disparate uses and architecture.
5. The Internet of Things Is a Thing of the Future.
No, actually, it’s a thing of the present. The technologies already exist. Connected objects call on hardware such as sensors, microcontrollers and microprocessors, as well as software such as communication protocols, operating systems and languages that are robust and widespread. New and innovative applications are emerging every day, though most are for industrial use and therefore unknown to the general public. Others are going to become so mundane that we’ll forget that they’re actually connected objects.
6. The Internet of Things is for Large Corporations.
Not necessarily. In fact, far from it. The concept of connected objects resting on open standards and technologies is inexpensive and does not require considerable financial wherewithal. In fact, it can be used to solve SME problems at a very low initial investment. Even the smallest companies can undertake such projects, as demonstrated by the numerous examples of crowdsourcing platforms. The Internet of Things is a vast innovation universe, and some of the innovations that have changed history started in a garage.
New technologies are always unpredictable, and the newer they are, the more we magnify the risks, real or imaginary. Many operators are working on making the Internet of Things a safe, reliable and interoperable world. This will happen through open standards and systems, and data security. Connected object solutions can be a decisive competitive advantage for businesses. Given all this, it would be irresponsible to simply ignore this new technological development just because of some of its inherent risk.